Human-machine interaction enables a user to communicate with a machine by speaking, typing, or performing gestures to express their interest, requirements, or queries. The human-machine interface may involve use of Embodied Conversational Agents (ECA) as animated characters or avatars to enable a meaningful interaction between a user and the interface.
However, the conventional human-machine interfaces fail to incorporate variations in human behavior, attitude, emotion, and sentiment during the interaction. In addition, the conventional interfaces provide a limited depth in expression while responding to the variations in the human behavior. Thus, the conventional interfaces are jittery in their expressions and interactions, thereby failing to provide an engaging conversation.
Moreover, the conventional interfaces fail to evaluate the human behavior based on context of the interaction. Also, the reaction of the ECA to an input provided by the user is pre-determined for the evaluated human behavior. Therefore, the interfaces are unable to provide a true emotional feedback and a dynamic dialogue.